Challenges in Quantitative Abstractions for Collective Adaptive Systems

TL;DR

This paper reviews recent advances and challenges in creating quantitative abstraction techniques for collective adaptive systems modeled by ordinary differential equations, aiming to reduce complexity while preserving system dynamics.

Contribution

It analyzes current aggregation-based abstraction methods, highlights their limitations, and identifies key challenges for future research in this area.

Findings

Aggregation techniques can simplify models but often lose important dynamics

Current methods face computational and accuracy limitations

Future research should focus on preserving critical behaviors in abstractions

Abstract

Like with most large-scale systems, the evaluation of quantitative properties of collective adaptive systems is an important issue that crosscuts all its development stages, from design (in the case of engineered systems) to runtime monitoring and control. Unfortunately it is a difficult problem to tackle in general, due to the typically high computational cost involved in the analysis. This calls for the development of appropriate quantitative abstraction techniques that preserve most of the system's dynamical behaviour using a more compact representation. This paper focuses on models based on ordinary differential equations and reviews recent results where abstraction is achieved by aggregation of variables, reflecting on the shortcomings in the state of the art and setting out challenges for future research.